There are overlooked scenes in an overlooked 100-year old sci-fi story that are becoming increasingly relevant at the dawn of humanoid robots.
What it might mean for robots to “feel” something deserves greater attention.
It may seem far-fetched. But, AI companies are working to collect more data from the physical world through a wide range of sensors, including tactile.
Could something resembling pain be closer than we think?
This week Ink & Time brings you:
Defining scenes from R.U.R. by Karel Capek on the topic of robot pain
A comparison of human infant learning and robot / AI learning
The latest stats in the US-China competition in humanoid robots
A special offer for ordering the Time Warp Editions of R.U.R.
AI companies are scrambling to capture more data to train their latest models.
It’s the real reason so much money is going into humanoid robots, in addition to the utility and and expected productivity gains, some of which is debated.
The biggest source of that additional data? The real world.
The challenge? Computers don’t have senses like we do, and have hit limits to what they can “learn” from all the text on the internet, which is insufficient.
Training on synthetic AI-generated text and images degrades performance. It’s sort of like incest for AI brains.
Thus, engineers are racing to put all manner of sensors on robots to collect real world data. Cameras. Mics. Sensors for touch and texture.
So, would robots ever feel pain, and know what that means?
That’s the turning point in R.U.R. Robots are given pain nerves to divert them away from mutilating themselves in machine work, obviously a cost to the world-beating company that manufactures fleets of them.
In a feat of wild tech speculation it triggers them to wake up, to develop sentience.
Given the vast sums at stake, expect a continued relentless push for more capable, embodied robots that can absorb as much data from the physical world as possible. Even if it means crossing the pain and suffering threshold.
If you read the dark conclusion in the R.U.R. story, you’ll know it’s not a good outcome.
Is it a risk worth gambling with? Big Tech investment thinks so. Maybe they never read RUR. Order a copy. Read it. Send a copy to your AI developer friends.
R.U.R. from Time Warp Editions, updated, annotated and illustrated (2024)
“We must introduce suffering…” A scene from R.U.R.
Excerpt from ACT ONE of R.U.R. by Karel Capek:
Hallemeier. …They’ve no interest in anything. Nobody’s ever yet seen a Robot smile.
Helena. Why—why don’t you make them—happier?
Hallemeier. That wouldn’t do, Miss Glory. They are only workmen.
Helena. Oh, but they’re so intelligent.
Hallemeier. Confoundedly so, but they’re nothing else. They’ve no will of their own. No soul. No passion.
Helena. No love?
Hallemeier. Love? Huh! Rather not. Robots don’t love. Not even themselves.
Helena. No defiance?
Hallemeier. Defiance? I don’t know. Only rarely, from time to time.
Helena. What happens then?
Hallemeier. Nothing particular. Occasionally they seem to go off their heads. Something like epilepsy, you know. It’s called “Robot’s Cramp.” They’ll suddenly sling down everything they’re holding, stand still, gnash their teeth—and then they have to go into the stamping-mill. It’s evidently some breakdown in the mechanism.
Domin. (Sitting on desk) A flaw in the works that has to be removed.
Helena. No, no, that’s the soul.
Fabry. (Humorously) Do you think that the soul first shows itself by a gnashing of teeth? (Men chuckle.)
Helena. Perhaps it’s just a sign that there’s a struggle within. Perhaps it’s a sort of revolt. Oh, if you could infuse them with it.
Domin. That’ll be remedied, Miss Glory. Doctor Gall is just making some experiments.
Dr. Gall. Not with regard to that, Domin. At present I am making pain nerves.
Helena. Pain nerves?
Dr. Gall. Yes, the Robots feel practically no bodily pain. You see, young Rossum provided them with too limited a nervous system. We must introduce suffering.
Helena. Why do you want to cause them pain?
Dr. Gall. For industrial reasons, Miss Glory. Sometimes a Robot does damage to himself because it doesn’t hurt him. He puts his hand into the machine— (Describes with gesture) —breaks his finger— (Describes with gesture) —smashes his head. It’s all the same to him. We must provide them with pain. That’s an automatic protection against damage.
Helena. Will they be happier when they feel pain?
Dr. Gall. On the contrary; but they will be more perfect from a technical point of view.
Is a robot more technically perfect if it can feel pain?
Human Infant Learning vs. AI Robotics: From Neural Development & Data Transfer Rates to Socially Embedded Learning
So, what are the differences between human and machine learning?
And how does this relate to wiring up robots with “pain nerves” or at preliminarily with tactile sensors that can train their AI brains?
The gap between biological and artificial learning remains profound.
Human infant brains demonstrate extraordinary data processing power. During peak development, babies create an astounding 2 million new synaptic connections every second, forming 1,000 trillion synapses by age 3, nearly double adult brain connectivity.
Explosive neural growth enables infants to process multimodal sensory information simultaneously from birth, with primary sensory networks achieving adult-like functionality within weeks.
Human babies master fundamental concepts like object permanence, causality, and motor skills through direct embodied experience with relatively few examples. Research shows that motor training at 3 months significantly affects cognitive abilities 12 months later, demonstrating how early physical interactions create lasting learning foundations through experience-dependent neural plasticity.
In contrast, current AI systems face massive data requirements that dwarf human needs. Robotics foundation models demand terabytes of training data to achieve basic competencies, with researchers estimating 6,377 years of continuous robot data collection would be needed to match language model performance levels. Modern frontier AI models require over 10^25 floating-point operations and cost hundreds of millions of dollars to train.
Current humanoid robots process sensor data at gigabits per second yet struggle with navigation tasks that toddlers master effortlessly. The persistent "sim-to-real gap" means that even systems trained on real-world data require exponentially more examples than humans to achieve comparable performance.
The fundamental difference lies in learning architecture. Human brains operate through continuous, analog processing with 60% of infant energy devoted to neural development, enabling simultaneous learning across multiple domains. AI systems rely on discrete digital computation and batch learning from static datasets, struggling with catastrophic forgetting when acquiring new skills.
Perhaps most critically, human infant learning is socially embedded and culturally situated, with caregivers providing emotional context and behavioral modeling that current robotics systems cannot access. This social scaffolding enables incredibly efficient learning transfer that artificial systems have yet to replicate, highlighting the profound challenges facing AI development despite impressive advances in real-world data collection capabilities.
For now humans as infants and at all ages, feel pain. Robots do not, yet.
“Perhaps this hatred is more like human beings”
From ACT TWO of R.U.R.:
Dr. Gall. Come here to the window, my good fellow. Let’s have a look. Please give me a needle or a pin.
Helena. What for?
Dr. Gall. A test. (Helena gives him the needle. Gall crosses up to Radius, who faces him. Sticks it into his hand and Radius gives a violent start.) Gently, gently. (Opens the jacket of Radius and puts his ear to his heart) Radius, you are going into the stamping mill, do you understand? There they’ll kill you— (Takes glasses off and cleans them) —and grind you to powder. (Radius opens hands and fingers.) That’s terribly painful. It will make you scream aloud. (Opens Radius’s eye. Radius trembles.)
Helena. Doctor—
Dr. Gall. No, no, Radius, I was wrong. I forgot that Madame Domin has put in a good word for you, and you’ll be left off. (Listens to heart) Ah, that does make a difference. (Radius relaxes. Again listens to his heart for a reaction) All right—you can go.
Radius. You do unnecessary things— (Exit Radius L.2.)
Dr. Gall. Reaction of the pupils, increase of sensitiveness. It wasn’t an attack characteristic of the Robots.
Helena. What was it, then?
Dr. Gall. (C.) Heaven knows. Stubbornness, anger or revolt—I don’t know. And his heart, too. It was fluttering with nervousness like a human heart. He was all in a sweat with fear, and—do you know, I don’t believe the rascal is a Robot at all any longer.
Helena. If you knew how he hates us. Oh, Doctor, are all your Robots like that? All the new ones that you began to make in a different way? (She invites him to sit beside her. He sits.)
Dr. Gall. Well, some are more sensitive than others. They’re all more human beings than Rossum’s Robots were.
Helena. Perhaps this hatred is more like human beings, too?
Dr. Gall. That too is progress.
The Great Humanoid Robot Race: China vs. United States
Whether we like it or note, the humanoid robotics industry stands at a pivotal moment. China and the United States are locked in a technological competition that will define the future of automation.
Not enough of us are paying attention and scrutinizing this race for the myriad risks it presents. Read R.U.R. in 2025 and it no longer feels like science fiction.
China currently dominates production with over 30,000 humanoid robot orders expected in 2025, nearly double the US projection of 18,000 units.
The global market, valued at $2.24 billion in 2024, is exploding to $7.8 billion in 2025 and projected to reach $38 billion by 2030. By 2035, analysts forecast a $181.9 billion industry, with Morgan Stanley predicting China will deploy 302 million humanoid units compared to America's 77 million.
China's strategy centers on scale and government coordination. Beijing allocated $20 billion to humanoid robotics in 2024 alone, establishing a $138 billion fund for AI and robotics over two decades. Companies like UBTech, which secured a $1 billion credit line, and AgiBot, planning a multi-billion IPO, with robots priced at $15,000-25,000.
The United States counters with innovation and premium positioning. Figure AI's remarkable $1 billion funding round in September 2025, achieving a $39 billion valuation, demonstrates American venture capital's confidence in breakthrough technology. Tesla maintains ambitious plans for one million Optimus robots annually by 2030. Agility Robotics operates the world's first dedicated humanoid production facility, already deploying robots in Amazon warehouses.
With analysts projecting a $5 trillion market by 2050, the nation that achieves humanoid robot dominance will command the next industrial revolution.
It’s a race to determine humanity's robotic future.
Read The Precursor to Asimov’s Three Laws of Robotics
The framework that features in many of his robot stories rightfully gets attention as humanoids start to appear on the scene.
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
What most people don’t know is that Isaac Asimov was born the year R.U.R. was published, and he took early inspiration from the book, amongst others.
He famously disagreed with the premise of R.U.R., but the corporate fleet model of robot workers is very similar to the direction of industry today.
Even if you think robot “feelings” is a crazy science fiction idea that will never come to pass, reading Karel Capek’s original humanoid robot story will give you much needed perspective on the next phase of embodied AI industrialisation.
Special Offer: Order a copy of Time Warp Editions’ R.U.R. from Amazon, send us a screen grab of your order, and we will send you another Time Warp e-book of your choice absolutely free!
Click to order R.U.R. in hardback or paperback
Ink & Time is pleased to bring you long forgotten works of literature, science fiction, art, philosophy and more that hold special relevance to today’s world.
Help us promote a new literacy in an attention starved world.